Head injury remains the leading cause of traumatic death, being responsible for 50-60% of fatalities. Of those dying of head injury, 66% have evidence of secondary insults due to reduced cerebral oxygen delivery. Most of these secondary insults are due to intracranial hypertension due to intracranial hematoma, swelling or vasogenic edema. Swelling and vasogenic edema can be exacerbated by intravenous fluid therapy which is necessary to treat hypovolemia and maintain a normal state of hydration. Recent laboratory investigations of head injury suggest that hypertonic fluid may adequately restore and maintain systemic hemodynamics while at the same time reducing intracranial pressure. If such results could be reproduced in patients with head injuries, hypertonic fluid could decrease the frequency of intracranial hypertension and lessen the incidence of secondary injury. We propose to compare a hypertonic fluid regimen to a hypotonic fluid regimen in patients with moderate and severe brain injury by determining the effects of fluid infusion on intracranial pressure. Specifically, we will compare hypertonic saline (1.6%) to a slightly hypotonic saline (0.7%) in the treatment of hypotension or intraoperative blood loss. For those patients without hypotension and who do not require operation, we will compare a hypertonic maintenance fluid (0.9%) to a hypotonic maintenance fluid (0.45%) fluid. Patients will be stratified by injury type and severity prior to randomization which will be done prospectively at the time entry criteria are fulfilled. Patients will be followed throughout the hospital course and at 6 weeks, 6 months and 1 year post discharge.